rscheffler wrote:
Edward, I understand where you're coming from. That is probably the ME upgrade in 2 years. Or one of the M240 product line expansion options. But too much to hope for right now as it would seriously undermine interest in the M240.
Fully agreed. I very well imagine Leica in a couple of years release a limited edition Typ 250 with pure RF experience, for 1000$ extra Perhaps Typ 260 Monochrom too
Edward, I think when supply of the M9 sensor runs out, my guess is that we will probably have a Leica ME.2 but with the new CMOSIS sensor without all the frills, now whether that is a good or bad thing ....
joe88 wrote:
Edward, I think when supply of the M9 sensor runs out, my guess is that we will probably have a Leica ME.2 but with the new CMOSIS sensor without all the frills, now whether that is a good or bad thing ....
I certainly agree with that. As long as it is priced like the ME that is
edwardkaraa wrote:
I have used the M9 for landscape photography on tripod and the results are too sharp with moire everywhere, so if anything I prefer it to be less sharp or less focus accuracy if anything
edwardkaraa wrote:
Also the improved RF accuracy and resistance to misalignment as well.
Resistance to misalignment would be nice. Though I only had that one time and it wasn't a bump or something like that. I was a repair guy (non leica) who aligned my good camera to a faulty lens he just screwed when coding it. Took me forever to find out why that lens worked perfect while my other lenses didn't.
I read a lot about misaligned camera's but can't really say its has been a big problem for me.
About the improved RF accuracy? Maybe I don't understand your post, but I can't remember the M240 has an improved RF? Does it?
For me the restrictions for the lifeview function of the M240 is clearly a deal breaker. I shoot 99% from a tripod and really need to be able to zoom on any point of the image. I just wrote an email to Leica CS and asked them for a statement, whether this restriction will be changed in the near future with a firmware update.
At the moment I'm #1 on the list of the largest Leica dealer in Germany for two M240. If I don't get at clear (and positive) statement, I will cancel my order and will even consider to change back to a DSLR system (D800E with Zeiss).
Maybe as a landscape photographer I'm not one of the typical core customers for the M, but I really like the small size of the system and the very high quality of the lenses. The rangefinder is just a nice add-on for me, but absolutely not necessary. I would probably even buy a Leica M with just a good lifeview implementation and no optical viewfinder at all (as long as it is full frame).
But as Dan wrote, if I cancel my order some other people will be happy to get their M240 a little bit earlier.
Boris - you are producing excellent results with the M9 which does not have LV at all. Am I missing something here? Please let us know what sort of response you get back from Lecia....
Wow, Boris, by Dan's statement you have now joined the ranks of Leica haters!
More seriously, this shows that Leica's market research either didn't uncover the likes of Boris, or the company not only doesn't really care about extending its market penetration to new users, but is also willing to see some existing customers walk away and sell their gear on the used market. And for what? Just not to offend some hard-core loyalists?
My guess is that the LiveView limitation is not hardware driven, but a self-imposed software implentation that can easily be reversed. And that Boris and others can continue their travels with Leica in their bags.
BTW, there's an ISO 6400 D600 comparison NEF file now available for the ISO comparison shots on LUF.
My take on it is the D600 file is about a stop over exposed, and the M240 ISO 5000 is about 1/3 stop under, so it's a difficult comparison. The luminance noise in the D600 file pulled back one stop is somewhat higher than the M240 ISO 3200 sample. If the D600 file is not pulled back, but the mid and shadow values are darkened in a curve adjustment, to somewhat match the look of the 1/3 stop pushed M240 file, it's pretty close. Chroma NR seems slightly better with the D600. With chroma NR off, the color noise of the M240 file is stronger. LR's default color NR setting of 25 works fine for the D600, but somewhere between 30-35 is needed for the M240 to match. Personally, leaving it at 25 is fine with me. With M9 files at ISO 1250 I was already often between 30-40 chroma NR and with ISO 2500 usually around 50, so it's a definite improvement. Keep in mind there still isn't proper M240 support in LR and future firmware revisions will likely change these observations anyway. I guess my ultimate hope is Leica can address the banding issue and completely eliminate it. If the luminance and color noise characteristics remain as they are, that'll be fine with me.
rscheffler wrote:
I guess my ultimate hope is Leica can address the banding issue and completely eliminate it.
Isn't banding caused by random sensor behavior? Can an updated firmware or software correct the randomness of this effect? Many users seem to think that the reason we were seeing banding on the M9 was due to the CCD sensor. Now with a switch to CMOS, why do we still have banding issues?
I don't currently own a DSLR anymore, so my current benchmark for FF IQ is my Sony RX1. The files from the RX1 are very malleable, with excellent dynamic range and combined with a very capable Zeiss 35mm lens, has resolving power that matches or nearly matches my M9 and Leica ASPH lenses at close to medium distances. If the M240 sensor can't compete with these current generation of Sony sensors, I would be somewhat disappointed.
joe88 wrote:
Isn't banding caused by random sensor behavior? Can an updated firmware or software correct the randomness of this effect? Many users seem to think that the reason we were seeing banding on the M9 was due to the CCD sensor. Now with a switch to CMOS, why do we still have banding issues?
I don't currently own a DSLR anymore, so my current benchmark for FF IQ is my Sony RX1. The files form the RX1 are very malleable, with excellent dynamic range and combined with a very capable Zeiss 35mm lens, has resolving power that matches or nearly matches my M9 and Leica ASPH lenses at close to medium distances. If the M240 sensor can't compete with these current generation of Sony sensors, I would be somewhat disappointed. ...Show more →
Even though CMOSIS is apparently a reputable sensor maker, I think it's optimistic to think that their sensors can ultimately compete with giant Sony's sensors. Canon can't even quite manage that, these days.
As long as the M 240's sensor is a noticeable improvement over the M9's, I'd say that's acceptable, although it remains to be seen what the differences in "look" will be. From what I've seen so far, it may be kind of a slide "film" vs. negative "film" type of situation.
Joe, I can't answer that because it's well beyond my area of knowledge. What I can offer, since this is an internet forum, is my 'gut feeling' about the M9. And also some experience with specific Canikon cameras.
One of my Canon 1DIII cameras had a one pixel wide white line that became progressively more visible as the ISO increased, but existed in every single frame. Canon replaced the sensor and it didn't resolve the problem. Finally they replaced the PCB related to the sensor, and that fixed the problem.
Closer to what we're seeing with the M240 samples, my 1D (original version) camera would show banding at higher ISOs. There was apparently a 'hardware fix' for that issue, but I got conflicting information from Canon Canada and never actually had it serviced. Based on forum commentary at the time, there wasn't clear indication that the hardware fix was related to replacing the sensor.
One of my newspaper colleagues shoots with a D300s that always results in across the frame pattern banding. He's oblivious to it, but I see it in all of his images. None of the other guys shooting D300s have that problem. One would think if it was a chronic sensor issue, everyone would see it. This also reminds me of the 1DsII - some guys reported having weird pattern banding issues with it, but it not an issue that affected everyone.
With the M9, my gut feeling is the sensor is probably OK, rather, it's the mishmash of components and potentially insufficient RF shielding at the root of the problem. I think pretty much everyone at some point experiences M9 image banding. But, if it was the sensor, then I think the banding would be consistent with all frames. Does the M8 also exhibit random banding? I really haven't kept track of pre-M9 discussions.
From the early M240 samples, there are definitely images where if you push them 3 stops, you don't see the banding, but then, those images also typically don't have areas of pure black in them where pushing more than three stops still digs into deeper densities. The other problem is we've really only seen images from 4-5 M bodies. I think Jono said he's used three, and Chris Tribble's samples are all from one Jono used. Then there are Dr. Rohde's samples and those from Jean Gaumy. But the latter's images are all 1000 pixel wide web-Jpegs, mostly shot in good light at low ISO, though a few are at ISO 1000.
FWIW, Ron, I agree that it doesn't seem to be the sensor of the M9 that has the banding problem, per se. I just did yet another banding test in C mode with my M9 in very low light, and found something interesting. Using either my Transcend or Sandisk card, if I hold the shutter down for five exposures in C mode, there is no banding in the first pic, the second pic you start to begin to see it, and then it gets worse from there.
If I keep the camera in C mode, shoot one pic, take a 3 count AFTER the red light is done flashing, and then fire again, there is no banding in the second. I tried this for a few shots is a row, and there was no noticeable banding.
Finally, I took one shot, waited for a 3 count after the red light, took a second shot, waited for a 2 count after the light, took another shot, waited for a one count after the light, took another shoot, and then didn't wait at all after the red light flashed, before taking the final shot. So? First two shots had no banding. third shot had very noticeable banding at the top 1/12th-ish of the photo. Fourth shot had noticeable banding in the top half of the photo. Fifth shot had noticeable banding across all of photo.
So, it seems to me that, if I don't want banding in very lowlight, I need a 3 count after the red light is done flashing before I fire another one. My first shot always seems to be ok.
p.s. these are ISO 160 shots pushed 4-5 stops with deep shadows, so it's pretty extreme.
Thanks Doug, very informative. It's similar to what I've also experienced with shooting off continuous sequences. Good to know about the 3 second interval.
Ron, I trust your "gut" feeling more than a lot of other review sites online
Douglas, interesting observation with the delay to avoid banding. I don't think I have the inclination to count like you do while out in the field especially when I'm shooting a scene that is unfolding, its hard enough for me to take decent pictures without doing something else but I can see how this technique could work in deliberate or planned shooting situations.
My best card experiences after shooting the M9 for over 2.5 years are with the 16GB and 32GB Class 10 Transcend cards. Although I still get banding with these cards occasionally, they occur less often. All other cards I used, SanDisk Extreme & Ultra, Lexar, Kingston, etc had multiple issues as documented everywhere, not only banding but lock up, extremely slow buffer, etc. Banding does get worse when the sensor is heated up after a few exposures. But like Ron, said it could be the entire circuitry design rather than sensor alone and if that is the case, I think it would be too late for Leica to rework the entire circuitry if they want to release the camera this year. And if some of us were thinking of shooting videos, the sensor would heat even more and banding could be worse.
On my now sold M8, anything above iso640 and I will have a banding feast, so the M9 is not an issue for me
BTW, even my Monochrom exhibits random banding when shadows are pushed hard.
Yeah, I did my test with both my Transcend class 10 16gb and Sandisk Exteme Pro 16gb, and both exhibited the issue. I've been using the Sandisk, since it writes much faster.
Granted, so many variables like battery power, camera settings, room temp., etc. are an issue, so who knows?
Something else I observed with the M9: I tested a Transcend 16GB Class 10 SDXC card and it was a disaster. The worst banding I've ever seen. This was the 'icing on the cake' that led me to believe its not the sensor but something else to do with how the files are processed or moved between the sensor, buffer, CPU and card reader/writer.
Well, it could still be the sensor, or components on the sensor, that are extremely sensitive to fluctuations or interference of some kind.
Joe, I couldn't agree more about the Sony sensor on the RX1 with the 35/2 lens. Low light is great at 6400 ISO, and has excellent DR. I would really hope the new M 240 is equal if not better. Only the caveat, is that at base ISO's, the IQ needs to be on par at least with the M9.
Very interesting info Doug and Ron! I have had a similar issue with the M-M, with 4 or 5 fast sequence shots, with also horizontal lines appearing. I sent these shots via Camera clinic, and the RAW files are with Leica Solms. The problem has been reproduced in the lab also. They have acknowledged the problem, and again it has to do with sequential shots, where the buffer has not cleared. From the head person, reviewing the files, he has suggested it is probably a voltage problem, and interference. I have since slowed down, and no problems with either the M9 or M-M for now.
Perhaps Typ 260 Monochrom too 
